There is a dearth of information on the mechanisms, and potential moderators, of physical activity (PA) intervention effectiveness. Understanding the mechanisms of PA change will aid in successful intervention development while understanding moderators and will allow a matching of intervention components to individual needs of particular populations. To definitively understand mechanisms and moderators of behavior change it is necessary to develop theory-based interventions that include specific strategies that target proposed mechanisms of change while monitoring the proposed mediating and moderating variables prior to and following intervention. This proposal highlights the use of the Ecological Model of Physical Activity (EMPA) as a theoretical guide to strategy development. The EMPA highlights the necessity to address individual (biological & psychosocial) and environmental variables (e.g., PA resource accessibility) when developing and implementing interventions. Health care patients who have been referred for a treadmill cardiovascular stress test--but do not present abnormal ECG or chest pain during testing-will be will be stratified by diagnosis of low heart rate recovery, a potential biological moderator of intervention effectiveness. To detect the potential independent and combined effects of individual motivation and environmental accessibility we will conduct a randomized 2X2 factorial trial to test proposed mediators of effectiveness. Participants (n=500) will be randomized to an individual (I), environmental (E), combination (l+E), or matched contact nutrition control. All intervention conditions will include a CD-ROM session, 3 tailored mail, and 3 tailored automated telephone follow-ups over a 6-month period. We will assess the reach, effectiveness, clinician adoption, and staff implementation of these interventions, individually, in combination, and when compared to controls among an understudied and high need population--individuals who are at an elevated risk, but have not yet suffered a serious cardiac event (approximately 3,000,000 individuals/year in the US). The results of this project will identify the causal, and potentially additive, relationships of environmental and individual mediators of PA change and allow the examination of potential biological moderators of effectiveness. Because intervention typically effects decrease over time and studies do not address setting-related issues necessary to ensure sustainability of delivery at the organizational level we will determine the maintenance of intervention effects on causal mechanisms and PA, 6 & 12 months following intervention. The interventions will be developed for delivery within the regular scope of treadmill stress testing by typical medical staff and cost of delivery will be assessed for each intervention condition. This project uniquely utilizes methods that will facilitate institutionalization of the intervention components following the successful completion of the trial and potential adoption of the intervention in other health care settings.